DESCRIPTION: (Applicant's Abstract) This study, entering its fourth year at the time of this writing, seeks to test, revise and improve the predictive validity of the Criteria of the American Society of Addiction Medicine (ASAM). Although now widely disseminated and endorsed by many payers, several states and the US Department of Defense, the ASAM Criteria have not undergone prospective testing. This study was the first, and to our knowledge, still remains the only one to undertake a randomized, double blind prospective match-mismatch design to test whether matching according to the ASAM Criteria is feasible and valid. The subjects are volunteer adult substance dependent treatment seekers who agree to be either matched according to the ASAM Criteria, or mismatched (within medically and ethically permissible levels of care that are within the community standard). Outcome measures consist of (a) addiction severity, (b) functional general health status and (c) health service utilization costs. Random regression modeling is used to test the hypotheses that outcomes are determined by either treatment intensity, or ASAM Criteria match, or an interaction between treatment and match. Results are promising, indicating that the ASAM Criteria: 1) can achieve a quite satisfactory inter-rater reliability; 2) do differentiate patients into levels of care that show expected differences along multiple clinical scales; 3) may not be easily accepted by patients without system supports that are sensitive and specific to subpopulations, and 4) may predict clinical, functional and utilization improvements in outcome. The specific aims of this competing renewal are to complete the originally planned subject accrual and revise the original ASAM Criteria based on these outcomes, and then use this complete data set for two new objectives: 1. to map, computerize and retrospectively test the reliability and validity of the second edition of the ASAM Criteria, comparing it to the original published version. This project's success in mapping and implementing the ASAM Criteria-l offer an efficient opportunity for extending this work to the ASAM Criteria-2. 2. to develop a new, fully unbundled model of matching services to needs (the Cumulative Block Increment model) retrospectively using the data set to test its validity against the original and second editions of the ASAM Criteria. At the conclusion of this project, the result should be an empirically-driven revision of the ASAM Criteria that is adequate for distribution to investigators in subsequent research trials and clinical demonstration projects.